Systems and methods for real time suggestion bot

Abstract

Disclosed herein are embodiments of systems and methods for automated real time exploration of topics of interest during an electronic communication session. One or more meeting participants identify a category of interest, and operate an electronic device in the electronic communication session. A processor executes a machine learning model to identify one or more spoken words within a set of spoken words during the electronic communication session as one or more units of interest corresponding to the category of interest. The machine learning model may be trained to determine a context of the set of spoken works and to identify the one or more units of interest based on the context. The processor retrieves content associated with the units of interest from one or more data collections associated with the category of interest. The processor presents the content for display in real time in a graphical user interface.

What is claimed is: 
     
         1 . A method comprising:
 identifying, by a processor, a category of interest to one or more meeting participants operating an electronic device in an electronic communication session;   executing, by the processor, a machine learning model to identify one or more spoken words within a set of spoken words during the electronic communication session as one or more units of interest corresponding to the category of interest,   wherein the machine learning model is trained to determine a context of the set of spoken words and to identify the one or more units of interest based on the context of the set of spoken words during the electronic communication session, wherein the machine learning model was previously trained based on meeting topics data and word topics data prepared before the electronic communication session;   retrieving, by the processor from one or more data collections associated with the category of interest, content associated with the one or more units of interest; and   presenting, by the processor for display on the electronic device in real time during the electronic communication session, the content.   
     
     
         2 . The method of  claim 1 , wherein the set of spoken words comprises a speech-to-text transcript generated via automatic speech recognition of an audio including utterances of the one or more meeting participants. 
     
     
         3 . The meeting of  claim 1 , wherein the processor displays the content as an overlay of a graphical user interface of the electronic communication session. 
     
     
         4 . The meeting of  claim 3 , wherein the processor displays the overlay of the graphical user interface in a plurality of graphically distinct content segments corresponding to respective units of interest of the one or more units of interest. 
     
     
         5 . The method of  claim 3 , wherein the overlay comprises an input element configured to receive a text string from the electronic device and display an automatic response. 
     
     
         6 . The method of  claim 5 , wherein the automatic response is automatically generated by a chatbot interface of the processor. 
     
     
         7 . The method of  claim 1 , wherein the machine learning model was trained before the electronic communication session by applying a hierarchical topic model to data extracted from one or more of a meeting agenda, meeting topics suggestions, a presentation, or a conference paper. 
     
     
         8 . The method of  claim 1 , wherein the content comprises a link to a website. 
     
     
         9 . The method of  claim 1 , wherein the retrieving content associated with the one or more units of interest employs a set of search resources received before the electronic communication session. 
     
     
         10 . The method of  claim 1 , wherein identifying the category of interest comprises storing the category of interest in memory in communication with the processor before commencing the electronic communication session. 
     
     
         11 . The method of  claim 1 , wherein the category of interest is selected from the group consisting of meeting subject, word topic, specialty search engine category, and vertical search engine category. 
     
     
         12 . The method of  claim 1 , wherein the one or more units of interest comprise one or more of a keyword, a key phrase, a concept query, and a topic model. 
     
     
         13 . The method of  claim 1 , wherein the one or more units of interest comprise a sample image associated with the one or more spoken words, wherein the retrieving content associated with the one or more units of interest employs a content-based image retrieval (CBIR) query. 
     
     
         14 . The method of  claim 1 , further comprising:
 receiving, by the processor from the electronic device, an input indicating a rating for the content; and   training, by the processor, the machine learning model in accordance with the input.   
     
     
         15 . A system comprising:
 an electronic device being operated by one or more meeting participants operating an electronic device in an electronic communication session;   a storage medium storing a category of interest to the one or more meeting participants a server in communication with the storage medium and connected to the electronic device via one or more networks; wherein the server is configured to:
 execute a machine learning model to identify one or more spoken words within a set of spoken words during the electronic communication session as one or more units of interest corresponding to the category of interest, wherein the machine learning model is trained to determine a context of the set of spoken words and to identify the one or more units of interest based on the context of the set of spoken words, wherein the machine learning model was previously trained based on meeting topics data and word topics data prepared before the electronic communication session; 
 retrieve from one or more data collections associated with the category of interest, content associated with the one or more units of interest; and 
 present the content for display in real time during the electronic communication session. 
   
     
     
         16 . The system of  claim 15 , wherein the server is configured to present the content for display in real time as an overlay of a graphical user interface of the electronic communication session within a graphic frame in which the overlay may be scrolled. 
     
     
         17 . The system of  claim 15 , wherein the server is configured to present the content for display in real time as an overlay of the graphical user interface in a plurality of graphically distinct content segments corresponding to respective units of interest of the one or more units of interest. 
     
     
         18 . The system of  claim 15 , wherein the category of interest is selected from the group consisting of meeting subject, word topic, specialty search engine category, and vertical search engine category. 
     
     
         19 . A system comprising:
 a non-transitory storage medium storing a plurality of computer program instructions; and   a processor of a first electronic device electrically coupled to the non-transitory storage medium and configured to execute the plurality of computer program instructions to:
 identify a category of interest to one or more meeting participants operating a second electronic device in an electronic communication session; 
 execute a machine learning model to identify one or more spoken words within a set of spoken words during the electronic communication session as one or more units of interest corresponding to the category of interest, wherein the machine learning model is trained to determine a context of the set of spoken words and to identify the one or more units of interest based on the context of the set of spoken words during the electronic communication session, wherein the machine learning model was trained before the electronic communication session by applying a hierarchical topic model based on meeting topics data and word topics data prepared before the electronic communication session; 
 retrieve from one or more data collections associated with the category of interest, content associated with the one or more units of interest; and 
 present the content for display by the second electronic device in real time during the electronic communication session. 
   
     
     
         20 . The system of  claim 19 , wherein the processor of the first electronic device is configured to present the content for display by the second electronic device in real time as an overlay of a graphical user interface of the electronic communication session within a graphic frame in which the overlay may be scrolled.